I have a 480Y/277 V, 3 Phase 4000A circuit of 11 parallel [4-500kcmil-4" pvc conduit] in a ductbank. Contractors have cut wires to fit instead on maintaining same length per NEC 310.4. Conductors per phase vary from 80 to 90 feet total length for a maximum difference of 10 feet. Assuming maximum future load of 3200 amps, is this difference in length a concern? Conductors are copper.
Parallel Conductors
- Thread starter kzahn
- Start date
- Status
charlie tuna
Senior Member
- Location
- Florida
you are going to have problems -- ten foot difference on a eighty foot run is gonna cause the load to shift to the shorter runs and at 3200 amps it will probibly exceed the rated amperage of the shorter wire....... we had to build a 4000 amp jumper from the power company's vault to re-feed a switchgear due to the failure of the vault stub. we ended up with eight 500mcm cu per phase. the wire was layed out on the vinyl tiled floor and some of the runs were cut wrong by the supply house --- three foot short. i didn't think alot about it -- but after we energized it and reached the peak loads to an office tower --- the short runs were drawing 580 amps, while the long runs were about 260 amps. that taught me a lesson. we built a plywood chase over the top of the cable run and installed fans to cool the wire. and you could feel the heat coming out the end of the plywood chase!!!
sandsnow
Senior Member
- Location
- Southern California
Check out these two threads:
http://www.mikeholt.com/codeforum/ultimatebb.php?ubb=get_topic;f=11;t=006272
http://www.mikeholt.com/codeforum/ultimatebb.php?ubb=get_topic;f=3;t=001602#000000
It is a code violation plain and simple. Ten feet is not even close. It can be corrected based on that alone. Remember there is no fudge factor in this requirement.
You can determine the actual current flowing through each paralleled conductor by a formula in the second thread above.
As your current approaches the capacity of the circuit the imbalance will become more of a problem as the shorter conductors will have more current in them due to the lower resistance.
Most people remedy this by making the longer conductors shorter to equalise them.
Another solution just occurred to me and I would like others to comment on it.
What about making the shorter conductors longer? Would you consider the splice as negligble? It does seem to enter in to the realm of a violation of 310.4(2) since the conductors are now not the same material from end to end.
I think it is better than having a ten foot differential in length.
http://www.mikeholt.com/codeforum/ultimatebb.php?ubb=get_topic;f=11;t=006272
http://www.mikeholt.com/codeforum/ultimatebb.php?ubb=get_topic;f=3;t=001602#000000
It is a code violation plain and simple. Ten feet is not even close. It can be corrected based on that alone. Remember there is no fudge factor in this requirement.
You can determine the actual current flowing through each paralleled conductor by a formula in the second thread above.
As your current approaches the capacity of the circuit the imbalance will become more of a problem as the shorter conductors will have more current in them due to the lower resistance.
Most people remedy this by making the longer conductors shorter to equalise them.
Another solution just occurred to me and I would like others to comment on it.
What about making the shorter conductors longer? Would you consider the splice as negligble? It does seem to enter in to the realm of a violation of 310.4(2) since the conductors are now not the same material from end to end.
I think it is better than having a ten foot differential in length.
I guess that if you used a copper split bolt (or compression) it would remove that violation, although I'm not sure that you wouldn't have to splice each conductor individually to meet the "terminated in the same manner" requirement. Also, the covering for the splice may not meet the "have the same insulation type" requirement. You might also require a JB depending on the type of installation.
Regardless, I don't think that splicing and adding to the short conductor is a good idea. If something ever happened to the splice and it became faulty or opened, you could have big (costly) problems in a hurry.
If my supplier shipped me conductors that I intended to parallel, and any one of them was too short because someone "mis-measured", I would send it back for one the proper length. I usually order one long length and make my own measurements and cuts. That way I only have one long (as short as possible) drop that I can maybe use somewhere else instead of several short pieces.
With the prices of copper these days, everyone needs to "measure twice and cut once".
steve
Regardless, I don't think that splicing and adding to the short conductor is a good idea. If something ever happened to the splice and it became faulty or opened, you could have big (costly) problems in a hurry.
If my supplier shipped me conductors that I intended to parallel, and any one of them was too short because someone "mis-measured", I would send it back for one the proper length. I usually order one long length and make my own measurements and cuts. That way I only have one long (as short as possible) drop that I can maybe use somewhere else instead of several short pieces.
With the prices of copper these days, everyone needs to "measure twice and cut once".
steve
- Location
- Illinois
- Occupation
- retired electrician
It may be a code violation, but it won't be a big problem. If 10 of the runs are 90' and one is 80', there will be ~400 amps on the 80' and ~360 amps on each of the 90' lenghts. The short run will be overloaded by 20 amps.
Don
Don
- Location
- Illinois
- Occupation
- retired electrician
Charlie,
There was something other than the length difference at work in your installation. The change in impedance and current is directly related to the change in lenght as a percentage of the total lenght of all of the runs.
Don
There was something other than the length difference at work in your installation. The change in impedance and current is directly related to the change in lenght as a percentage of the total lenght of all of the runs.
Don
charlie tuna
Senior Member
- Location
- Florida
there wasn't anything scientific about the jumper arrangement. eight runs for each phase layed side by side on the floor. we didn't know about the short runs until we were almost finished with the install and it was after hours. the short runs were humming from the current flow...............
hockeyoligist2
Senior Member
- Location
- close to greenville sc
We ran into a problem with that a few months back. we bought a roll of 500 and it was marked in feet. The boss wanted to cut it in 6 equal parts and ignored the markings..... wanted to use a tape......... needless to say the last one was about 15 feet short! Most of the big stuff is metered by a machine and is pretty accurate. When you are measuring in the field a few curves in 100 foot + sections can make a big differance!hillbilly said:
sandsnow
Senior Member
- Location
- Southern California
kzahn said:
That doesn't make it all right. You have to figure the loading to be sure.
I believe the reason there has never been a problem is because the actual loading is less than what is on paper.
charlie tuna
Senior Member
- Location
- Florida
from what i have seen the problem arises when you subject the feeder to it's full capacity. the closer you get to the rated wire capacity the larger the shift in current.
- Location
- Illinois
- Occupation
- retired electrician
Charlie,
Don
If you have all of the A phase next to each other and then all of the B phase and then the C phase conducotrs, the inpedance change caused by the inductive reactance will be much greater than the change caused by the length. You need to install these cables in bundles with a conductor from each phase in each bundle.
Don
- Location
- Illinois
- Occupation
- retired electrician
kzahn,
Don
In that case, I think that you have a problem....if the full load is 4000 amps, the 90' lengths will carry about 283 amps and the 80' ones about 460 amps.
Don
- Location
- Lockport, IL
- Occupation
- Semi-Retired Electrical Engineer
sandsnow said:
BINGO! Two Brownie Points for Larry.
I cringe whenever anyone says that it's never been a problem before. They're just not thinking the situation through.
- Location
- Wisconsin
- Occupation
- PE (Retired) - Power Systems
charlie b said:
Has any contractor ever said " I do it that way all the time, and it has always been a problem".
- Location
- Illinois
- Occupation
- retired electrician
Carl,
I don't know what I did the first time, I did it again and came up with 387 for the 80' ones and 344 for the 90' ones.
Don
Using your numbers there would be 6 at 340A and 5 at 383A.
I don't know what I did the first time, I did it again and came up with 387 for the 80' ones and 344 for the 90' ones.
Don
- Location
- Massachusetts
sandsnow said:
Funny thing is that no installation of parallel conductors meets the actual code requirement as written.
Not possible, every paralleled installation has different length conductors by some margin.
As written the NEC does not allow a 64th of inch difference of length.
IMO the NEC should provide a tolerance as a percentage of total length.
charlie b said:
Your assuming they did not think it through, the person might be like Don or Smart $ and has thought it through.
jim dungar said:
LOL
No probably not and certainly not all have thought it through.
I will admit that I do not worry that much about the equal length issue. I have not ended up with 10' difference in a 90' run that seems excessive. This issue needs to be considered when running the raceways to try to get it as close as possible.
However I do not stretch out sets of 500 or 600s on the floor to match up their lengths. We pull and terminate.
We do it all the time and it's never a problem
I think we touched on the reason it is not a problem, often the conductors are over sized for the load so there is some room for unequal loading.
- Status